Hitachi 2TB Harddrive Owner's Thread

odditory

Supreme [H]ardness
Joined
Dec 23, 2007
Messages
6,482
7K2000.jpg


Discussion of the Hitachi 2Tb HDS722020ALA330 harddisk.

A Wiki for this drive has been created HERE and includes firmware update tools and instructions.
 
Last edited:
If you haven't yet downloaded the "Hitachi Feature Tool", you may want to. It's a DOS based utility for changing various features of Hitachi drives, things like APM settings, SATA speed (SATA150 / SATA300), etc. It's available on the Hitachi downloads and utilities page here (the page also has tools like the DFT - Drive Fitness Test, etc):

http://www.hitachigst.com/hdd/support/download.htm#FeatureTool

Now I've read reports that HDS722020ALA330 drives manufactured prior to a certain date were set to "SATA 150" from the factory. Among other features, the Hitachi Feature Tool allows you to set the mode to "SATA 300". If you haven't bothered to check your drive for which mode it's set to, it might be a good idea. I'm verifying my drives now because in a raid array I wouldn't want some of them at SATA150 while others are at SATA300.

EDIT: Verified all my drives (manufactured October, November and December 2009) and all were already at SATA 300.
 
Last edited:
Working fine except for the bad port #8 that my 1680ix-24 developed- its bizarre. Switched cables, backplane, tried different drives, checked miniSAS connector #2 for debris, updated to latest firmware, everything. I've narrowed the problem down to the card simply refusing to deal with anything on Port8, so i'm trying to get it swapped. That's another discussion though.

I did attach the drives to my Areca 1280ML as well as my Adaptec 52445, both which had no problems building the RAID6 array- it was fast. I ordered two HP SAS expanders, and longterm I'm going to attach the drives to the expanders that will be driven by an Adaptec 5805.

EDIT: Adaptec 5805 did *not* work with HP expander, however several Areca 1680 series cards did.
 
Last edited:
Getting better speeds with your 1280ML or 1680ix-24? I own both at the moment, so I'm just a bit curious.
 
Have 4 of these in RAID5 on Areca 1231ML. Plan to expand to 6 in Late Jan/Early Feb. Then to 8 in April or so. (RAID6 @ 8).

No dropouts, etc. I was actually interested in building a 24 drive WD20EARS box for the Advanced Format, but I fear the dropouts. Enterprise drives too expensive, so I'll likely continue purchasing Hitachi 2TB drives.
 
Any ideas if Hitachi will announce anything new at CES? I'm in the same situation a lot of people are with these seeming to be the last option for 2tb drives in RAID arrays, unless anyone's tried the samsung f3eg 2tb drives.
 
A slightly long comment on odditory's note on Hitachi drive.

That's the reason why there is a slight difference between Hitachi and other brands.

If Hitachi is still at the core .. , and I sure hope they still keep this spirit decades later, a traditional "Japanese-culture" enterprise, they will value certain properties ingrained in their culture, the honour, the quality attitude, the desire to keep reputable products and promises to their demanding loyal customers,...sorry if this sounds a bit of like sales talk, I am not japanese I can assure you.

If you wanted one example how their differ in culture, Hitachi (as a mega-group of companies) loss untold billions over the years, yet they will make every attempts to keep employee, with pay/benefits cut, but nevertheless employed. I do not know how Hitachi has changed recently, I wish them best luck.

But how does this relates to hard disk? Hitachi is in 2 related businesses, the most premium Hitachi Data Storage (HDS) that sells to top to bottom enterprise market and the mainframe market, and the Hitachi Hard Disk business that supplies hard disks to disk buyers. (this group has assimilated other disk businesses over the years). It is in their best interest and culture to keep product quality and delivered features intact so as not to tarnish the reputation. Ask forum member SinShiva, I am happy Hitachi shipped a product as advertised in specification and not pull a Houdini act like some other choices' half-way manoeuvre.

I understand they bought the "Deathstar" business. I have no comment on other things except the above.
 
Last edited:
Any ideas if Hitachi will announce anything new at CES? I'm in the same situation a lot of people are with these seeming to be the last option for 2tb drives in RAID arrays, unless anyone's tried the samsung f3eg 2tb drives.

Someone replied in my other thread that they are using the 1.5TB F3EG drives in RAID no issue. At least on the same controller as me.

I'm interested in either these Hitachi drives or the Samsung F3EG 2TB for an array. I'd prefer the F3EG's for power consumption and lower platter count, but the Hitachi's seem to get good reviews in terms of RAID compatibility.
 
I just got 5 of them in a Raid 6 array. Initialized no problem and I am now doing a raidset/volume expansion. (http://www.hardforum.com/showthread.php?t=1481766)

I have also heard many of the horror stories with the current Seagates and WD's due. My previous array was made of Seagate 300GB 7200.9 These hard drives performed flawlessly for 3.5 years with never as much as a hiccup.

I hope I made the right choice with these Hitach 2TBs but I am glad that others have followed the same road.
 
Are there any kind of benchmarks on read/write performance? How does the drive compare speed-wise to a Samsung F3 1TB?
 
If you haven't yet downloaded the "Hitachi Feature Tool", you may want to. It's a DOS based utility for changing various features of Hitachi drives, things like APM settings, SATA speed (SATA150 / SATA300), etc. It's available on the Hitachi downloads and utilities page here (the page also has tools like the DFT - Drive Fitness Test, etc):

I'm just Advanced checking my 3 drives that came today, long process, going to build the array tomorrow and stress test it with BST5 for a few days I think, see if it drops out.

I was looking at the Hitachi Feature Tool, mainly to change the APM to the lowest setting. Not sure if APM will affect the reliability of a RAID 5 array. Anyone know? I was hoping just to set it as low as possible as I'm not really concerned with performance.
 
@Sircyro what are you connecting the drives to? A hardware based raid or an HBA + software raid? Reason I ask is because many hardware raid cards let you configure APM for the array (i.e. my Adaptec 5 series and Areca 1680 series cards both too), and so I much prefer setting APM on the controller than going one by one setting the drives with the tool.

I happen not to bother with the lowRPM or lowpower modes offered by the Hitachi, I just have array spindown configured @ 60 minutes and that's enough for me. In the future I may try experimenting with having the drives go LowPower or LowRPM after 15 minutes and see if based on my usage patterns that it would make any diff.
 
I'm running a RR4320, and I have my spin down time set to 10minutes. I build the system as low power as I could, so I'm just trying to keep up with the power savings.

I'd like to run the drives in LowPower or LowRPM I'm just not sure if that will cause the drives to not work in the array.

Perhaps I'm confused about how the APM works, but from what I've read in the Feature Tool white papers, I can set the drive to be on LowRPM or LowPower at all times while it's on? Or does it just change the timing of when that would occur?
 
I am seriously considering moving to this drive after 6 months of fighting with Seagate drives. I'm have my array on an Areca 1231 and am just about to RMA my 4th drive in those 6 months (from an array with only 8 drives no less - 50% hard failures in 6 months). My bad for failing to visit these forums and failing to learn from others mistakes...

Trying not to repeat that mistake, I am reading all I can before sinking more cash into the array. There seems to be a good deal of consensus liking the Hitachi 2Gb drives, but I'm also seeing lots of reports of these drives shipping DOA and poor random read/write performance (reported as so bad they suspect the cache is inop - read in a dozen or so reviews on newegg and other vendor sites). At least one users at AVSforum reported 2 of 8 DOA. Should I worry?
 
5 platters huh... No wonder it's the cheapest 2TB drive I saw today at CompUSA. They actually had two price tags below it (same exact SKU), one read $183 or $180-something and the other $205 IIRC. The Seagate/WD drives were $40-80 more. Incredibly enough they had absolutely no other 3.5" SATA drives between 2TB and 250GB, people must've hit 'em hard during the holidays. Edit: I'm aware they're much cheaper online ($160 at the 'egg), but my sister needed a new one ASAP, haven't bought anything yet tho since she hasn't decided if she wants another 2.5" external or a larger/faster 3.5".
 
I have a few of these running in various systems.

Pro is that they are cheap and appear to be reasonably quick. They make noise, but I would not call them noisy.

Con - they run very hot. If you are used to a WD Green drive, well, these run much, much hotter, make a lot more noise, yet don't seem to be particularly faster. So as someone who owns quite a few 2TB drives from Hitachi and WD, I can see why the WD drives carry a price premium. (I would never put the Hitachi in a HTPC or an external USB enclosure.)

I will say also that the drives just don't strike me as particularly fast.
 
I am seriously considering moving to this drive after 6 months of fighting with Seagate drives. I'm have my array on an Areca 1231 and am just about to RMA my 4th drive in those 6 months (from an array with only 8 drives no less - 50% hard failures in 6 months). My bad for failing to visit these forums and failing to learn from others mistakes...

What Seagate model do you have this problems with?
 
FYI, Microcenter has the retail Hitachi 2TB for $150 right now.

I bought 5 mid December from TigerDirect (retail packaged) for $130 after $30 rebate. Unfortunately 5 was the max. I'm checking TigerDirect and Newegg (they had the same deal in November which I missed) every so often to see if the deal pops back up.

I have the same experience as Odditory with the 1TBs. I have 12 of them in a Promise Technology iSCSI array device when they were first released. That was 3 years ago and they're still going strong with not one failure and that's in a garage that gets up into the high 90s in the summer. My Windows Server with 15 Seagate 1TB drives (Enterprise ES.2 drives to boot) have failed here and there over the past 2-3 years. I think I may have sent 5-6 back in that time. At least Seagate has been really good about swapping them out quickly under warranty. I heard Hitachi's warranty replacement procedure is quite slow.
 
So, has anyone managed to get APM to work on these drives, e.g. slower RPM? I mean, theoretically, if you could get them to spin at slower RPM (5400), it would be like having a green drive, but I've never had a SATA drive that actually worked with this feature.
 
7200.11 1.5Gb. The infamous ones. If I had just spent a bit more time reading first I could have avoided the headache.

Just wondering, are you having the firmware bug? I had one that was DOA, one that had the bad firmware and died week 1, but all 15 of them since have been flawless. These WD Green 1.5TB's however... a third (this week) of eight threw a SMART error last night. Alas, I've purchased 6x 2TB Hitachi's and 2x 1TB drives in the last week. They just work apparently.
 
Just wondering, are you having the firmware bug? I had one that was DOA, one that had the bad firmware and died week 1, but all 15 of them since have been flawless. These WD Green 1.5TB's however... a third (this week) of eight threw a SMART error last night. Alas, I've purchased 6x 2TB Hitachi's and 2x 1TB drives in the last week. They just work apparently.

No. All of my 7200.11 drives are firmware CC1H or CC1J which is the 'fixed' firmware. I have 20 of them. 8 in a WHS system behind AOC-SASLP-MV8 HBAs. 4 more in an older external enclosure from 3-ware (their 'sidecar') using a 9650SE contoller, raid 5. And 8 of them in a small raid-6 array with an Areca 1261ML-2G.

Interestingly, it is only the drives with the Areca controller that have ever failed. They drop out of the raid array regularly and every so often they pick up hard sector errors. 4 times they have gotten so bad that they could not be repaired using Seatools. The drives in the Sidecar/3ware are rock solid (but the 3ware controller is dog slow too). WHS and the Supermicro HBAs are happy as they can be - which is surprising since they sit in the garage and reported 40-45c temps most of last summer. I have no idea why they just hate the Areca raid controller. I guess cooking them in the garage is OK but running them at high performance is not. Seems backwards.

I am considering replacing those 8 with the 2Tb Hitachi drives but am concerned about the DOA reports and various problems with random read/write performance that have been reported.
 
I've been using a "dog slow" Adaptec 31605 with them... and for whatever reason they just work. The Hitachi's work. Everything just works, which is really nice. Plus, the old WHS only had two Realtek GigE's to saturate so even "dog slow" is easily fast enough.

I was thinking of putting the Hitachi's in a 'sidecar' but I'm super worried about something like an ant shorting out he external enclosure and losing the array. Probably highly irrational although my other idea is to just flip the Norco RPC-470 that has trayless backplanes and mount it atop the RPC-4020 making an oddly cobbled together 8Uish monster. I think it's a bad idea but I sketched it on a post-it in a meeting today.
 
I've been using a "dog slow" Adaptec 31605 with them... and for whatever reason they just work. The Hitachi's work. Everything just works, which is really nice. Plus, the old WHS only had two Realtek GigE's to saturate so even "dog slow" is easily fast enough.

I was thinking of putting the Hitachi's in a 'sidecar' but I'm super worried about something like an ant shorting out he external enclosure and losing the array. Probably highly irrational although my other idea is to just flip the Norco RPC-470 that has trayless backplanes and mount it atop the RPC-4020 making an oddly cobbled together 8Uish monster. I think it's a bad idea but I sketched it on a post-it in a meeting today.

can your replace those HDD bays with hot swap bays in the RPC-470?
 
I got 3 of these on Monday and have had good experiences with them so far. I ran the Hitachi Drive Fitness Test Quick/Advanced tests on all three drives, no issues. I hooked them up to my RR4320 and the RAID5 array built with no problems. I've been running BST5 on them going on 36 hours now. Two BST5's on the server itself, then several connecting from clients on my network, doing sequential & random read/writes simultaneously. No dropouts logged. I'm not sure how much longer I should keep the BST5 testing going. I stopped for a bit to do a few reboots and tweak some things on my controller.

Two issues I noticed:
1) With staggered spinup the drives would take a little too long for the RR4320 to see them all on boot, but within a couple seconds they were initialized fine and the array was normal. I disabled staggered spinup on the RR4320 and so far the array seems to boot normal right away now. I'm going to play around with the settings in the RR4320 a bit more, maybe setting it to more than one drive at a time staggered will help.

2) I don't think the drives are spinning down properly. I have my RR4320 set to spin down at 10 minutes and after say 20 minutes if I access the array I don't hear the drives spinning up. Also, I've been watching my kill-a-watt for the power readings, wattage doesn't drop until about 30 minutes or so after but I'm not sure that's even the drives as there is still no spinup time when accessing the array. Is there an APM setting I need to change to get these drives to spin down properly?
 
After fighting with the infamous Seagates for 6 months I have had enough. Did my 4th RMA out of an 8-drive array last week. I bit the bullet and bought 8 of these little Hitachi beauties when Buy.com put them on sale for $159. Finally got around to building the array - and I am impressed.

Areca 1261ml-2g, 8 drives, raid 6.

Initialization was faster than any other I've done of this size. Less than 4 hours - don't know exactly because I left it running not expecting it to finish for several more hours. When I stopped to check on it was surprised to see it done.

Tested with HD-tune. Average 510MB/s read, 350MB/s writes. A tad faster than it was with Seagate 7200.11s. Loaded it up and drove it as hard as I could, drives averaged 30-32c (the Seagates sat around 40c in the same case). And it was quiet - I used to hear the Seagates chatter (and occasionally that 'clunk/click/click' sound that you just dread...).

I think I'm in love. These things are the real deal for raid.

Noticed they are $154/each at Frys this weekend - limit 2. I may be making several trips...
 
@PigLover, good to hear. I suggest you only buy as many as you need, they'll go back on sale again for less than $150 soon enough. Not surprising that your Areca built the array fast. My Areca 1680 built a Raid6 with 20 Hitachi 2Tb's in about 14 hours.
 
I just looked before I started working this morning. The Hitachi drives were nicely spun down on my system (I have the Areca controller set to spin down after 30 minutes idle with 1 second spin-up stagger). Started up HD-tune and ran a read test and saw the expected 10 second or so delay before it started showing results. I don't have a kill-a-wat in line right now so I couldn't see the power usage.

One note: Unlike the Seagates, which clunked and clicked and generally sounded like they were about to die when the array spun up, the drives were nearly inaudible during spinup.
 
What makes these Hitachis work so smoothly anyway? Just better engineering or firmware? You'd think w/more platters on 'em they'd be more fragile, noisier, and/or warmer... I know it was forever ago but I still have bad memories of my dead 5-platter IBM 75GXP (I think that was the last 5-platter drive I've ever owned too). I admit I haven't kept up enough w/the HDD market, just enough to know what to avoid (like Seagate's 7200.11, etc.).
 
I'm running one of these in my Shuttle, and will add myself to the satisfied list. It spins up and down exactly as it should, and has never spit a single error. I actually didn't realize this was a 5-platter disk, and upgraded to it as a replacement from a set of 7K1000 drives (original 5-platter revision)- entirely coincidental. My fileserver has/had Seagate 7200.11's, all with good firmware, and yet they keep dying because of motor failure; having bought three originally, one died, as did its replacement, and then a different one most recently.

I will say that the Hitachi is REALLY loud, though- I can hear it over every other fan in my Shuttle, meaning a 1300rpm CPU fan, an 800rpm Scythe Kaze Slim and whatever the dinky fan is in the 300W PSU. It's my one gripe, but having it spin down after 30 minutes makes it livable.
 
The odd thing is that at the same time they introduced the 5 platter 2TB drives they introduced 2-platter 7200rpm 1TB drives, so they have the capability to make 4 platter 2TB drives. My guess is they're waiting to introduce those at a time when they announce 3 or 4TB drives later this year.
 
The odd thing is that at the same time they introduced the 5 platter 2TB drives they introduced 2-platter 7200rpm 1TB drives, so they have the capability to make 4 platter 2TB drives.

This is incorrect- if they could've put out a 4-platter 7200RPM 2Tb instead of 5-platter, they would have. There was an engineering issue specific to four platters @ 500Gb per platter when spun higher than a certain RPM. At 7200RPM supposedly the poor SNR produced error rates that were out of control. Meantime read/write head tech may have advanced and overcome that, but I don't think it was a marketing conspiracy if thats the concern.

Experience has shown the issue of 5 versus 4 platters ends up being more psychological than anything - 20% more platter and heads might lead you to believe its 20% more likely to fail but in real world use the failures just aren't happening. We have scads of every kind of 1Tb, 1.5Tb and 2Tb drive in large arrays at work and its the Seagate 1Tb ES.2 enterprise drives I see fail the most. One a month at least. Hardly a scientific study, but then again my 24 x 5-platter 1Tb Hitachi's have been going 24x7 for 2 years without even a single bad sector / media error reported in SMART, let alone drive failure, so I'm optimistic about the longevity of my 2Tb hitachi's.
 
Last edited:
What makes these Hitachis work so smoothly anyway? Just better engineering or firmware? You'd think w/more platters on 'em they'd be more fragile, noisier, and/or warmer... I know it was forever ago but I still have bad memories of my dead 5-platter IBM 75GXP (I think that was the last 5-platter drive I've ever owned too). I admit I haven't kept up enough w/the HDD market, just enough to know what to avoid (like Seagate's 7200.11, etc.).

No offense but having a bad experience with an IBM deathstar 5 or 10 years ago has zero in common with drives manufactured today, the one key factor being Hitachi took over production a long time ago, and with that came improved design, manufacturing, and >Q&A< which are obviously key..

Also I'm not sure anyone is trying to imply the Hitachi 2Tb is all out superior to other 2Tb drives on the market, its just that relative to the games some of the other manufacturers have been playing (like WD crippling TLER out of newer firmware, and some other brands not working right in raid arrays), the Hitachi has become kind of a defacto choice, especially for RAID use. Couple that with the decent pricepoint the Hitachi's have had during their time on the market and its only helped narrow the choice. Fry's had them on sale for $129 during the last week in December.

Now if a friend was building a WHS box and let's say the WD GP drives were $120 and the Hitachi was $140 and they asked "which one should I buy 8 of, and I'm on a budget" i'd probably tell them to go WD GP because its cheaper and they don't need raid.
 
This is incorrect- if they could've put out a 4-platter 7200RPM 2Tb instead of 5-platter, they would have. There was an engineering issue specific to four platters @ 500Gb per platter when spun higher than a certain RPM. At 7200RPM supposedly the poor SNR produced error rates that were out of control. Meantime read/write head tech may have advanced and overcome that, but I don't think it was a marketing conspiracy if thats the concern.

I was referring more to their platter's aerial density manufacturing capacities, and since they've jumped from 5 to 3 platter designs in the past surmising that since they've missed the four platter boat yet again they'll skip to three platters at whatever density they use for this year's drives (they scheduled the innovations necessary to give us at least 1TB/platter drives in 2010 after all, so maybe even a 2-platter 2tb design)
 
Experience has shown the issue of 5 versus 4 platters ends up being more psychological than anything - 20% more platter and heads might lead you to believe its 20% more likely to fail but in real world use the failures just aren't happening.

But it is still more power and more heat?
 
Expanded from (4) drive RAID5 to (6) drive RAID6. How long it took?

90:25:08 For Migration.
14:18:55 For Initialization.

On a 1231ML by the way. Plan to go to (12) drives eventually. I'll probably have another stop gap expansion in Feb or March. Never got a DOA or any death rattles yet. :p
 
Back
Top